Even Odds

[ddarko]

Well if you're going to bet every spin  - you will get the expected results.

I see it a different way, if your betting on say half the spins, your halfing your chances to win ?

& yes, lose.....

I'm not knocking the method, it just seems half complete to me......

[atlantis]

I see it a different way, if your betting on say half the spins, your halfing your chances to win ?

& yes, lose.....

I'm not knocking the method, it just seems half complete to me......

OK. Fine. In that case to get more bets you can use it applied to the other ECs: R/B and H/L as well......
Use separate banks for each EC?



A.

[ddarko]

OK. Fine. In that case to get more bets you can use it applied to the other ECs: R/B and H/L as well......
Use separate banks for each EC?



A.

That could well be an option  nice out of the box thinking !!!!

[thelaw]

Ran a few short tests (nothing definitive) using the following betting patterns :

EEE-bet O
EOE-bet E
EEO-bet E
EOO-bet E
OOO-bet E
OEO-bet O
OOE-bet O
OEE-bet O

Looks like 25% win across the board flat betting

[atlantis]

Hi thelaw,

Remember it is continuous tracking and betting - not just recording sets of lines of three results...

I'm chuckling a little coz in some strange way it reminds me a bit of that crazy iceman/winwithmath/wendell hidden math dimensional idea from years gone by. 


A.

[thelaw]

Hi thelaw,

Remember it is continuous tracking and betting - not just recording sets of lines of three results...

I'm chuckling a little coz in some strange way it reminds me a bit of that crazy iceman/winwithmath/wendell hidden math dimensional idea from years gone by. 


A.

Do you mean that after say EEE-bet O then you re-track (so if we had EEEE we wouldn't bet the second EEE pattern after the first loss)?

I ran continuous betting based on last three results (so 100 spins=97bets-3 to start tracking)

[atlantis]

Do you mean that after say EEE-bet O then you re-track (so if we had EEEE we wouldn't bet the second EEE pattern after the first loss)?

I ran continuous betting based on last three results (so 100 spins=97bets-3 to start tracking)

No - I believe you tested it right in accordance with the pdf file psimoes posted.
As I look at it, He doesn't seem to stop after the first loss of such a pattern you mention under the L3A+L2A heading.
However, the difference is you would not get anywhere near 97/100 bets testing O/E under his new idea because of the "no bets" from when L3A and L2A don't match.

A.

[thelaw]

No - I believe you tested it right in accordance with the pdf file psimoes posted.
As I look at it, He doesn't seem to stop after the first loss of such a pattern you mention under the L3A+L2A heading.
However, the difference is you would not get anywhere near 97/100 bets under his new idea because of the "no bets" from when L3A and L2A don't match.

A.

I have started to see that a run of EEEEEEE or EOEOEO can add up a large number of losses, which may mean a deep draw-down for the Bankroll

[atlantis]

I have started to see that a run of EEEEEEE or EOEOEO can add up a large number of losses, which may mean a deep draw-down for the Bankroll

Yes - right. I agree we need some guidance on that. We need to see what psimoes proposes is best to do in such an event...
Ideas, anyone??

A.

[atlantis]

The first thing that springs to mind:

If we get something like:

EEEEE or OOOOO it means we have 2 consecutive L's in our registry. (LL)

Same with:

OEOEO or EOEOE

as thelaw mentions.

We could STOP betting until pattern changes eg:

EEEEE (LL- stop) EEEO (restart here)

or alternate between L2F+L3F and (bet FOR the match)  and L3A+L2A eg:

EEEEE(E) delivers a W after the 2L's - now switch back to L3A+L2A.

If the run continues you will get alternate W's and L's eg:

EEEEEEEEEEEEEEO = LL(first 2 losses) then WLWLWLWLWW

instead of a larger string of losses. Not too bad!

If the pattern immediately ends after the 2 initial losses then the unwelcome streak ends anyway and we have a new pattern formed after our 3 consecutive losses eg:

EEEEEO = L-L-switch-L-switch

Therein lies the quandary to ponder; STOP after 2L's or use switch betting after 2L's until streak pattern breaks...

A.

[psimoes]

WOW thank you all for the nice words!

Atlantis, you understood the method correctly and brought new ideas to the table. Special thanks for that!

thelaw, your observations are right. It does lose when facing a long streak or a long series of chops.

In all honesty, I tried all possible combinations in that session. L3F+L2F, L3F+L2A, L3A+L2A, L3A+L2F and L3A+L2A achieved the best results for THAT session, as you can see in the pdf file. So I curvefitted the results. It it was just a lucky session and bad results from other sessions will prove it, my choice of L3A+L2A goes down the drain. Sorry for doing one test only and jumping to conclusions too early.

L3F+L2F will profit from those streaks and chops all the time. But that doesn't mean it will keep on winning for ever, as random will eventually catch up and there will be series of ECs that will beat L3F+L2F.

Curiously, L2F alone is what my variation on my favourite EC bet selection, the Author's system, is all about.

Well, what can we do to stop losing runs? Atlantis ' idea of a safety break is good, as well as changing from L3A+L2A to L3F+L2F after a series of losses. We don't even need to identify which patterns are going on. The Ls justify the changing of direction.

I thought about a "Control Group" consisting of betting Same As Last Spin at the same time as L3A+L2A.
Now there would be L3A+L2A+L1F! It's not getting any simpler, is it? In the case of a long streak of EEEEEEEEEE where L3A+L2A would both agreed with betting against, thus losing, L1F would contradict both. As there wouldn't be unanimity the conclusion would be No Bets. This would avoid long losing streaks.

Anyway, if we look at L3A+L2A or L3F+L2F we soon will see there will only be a limited selection of patterns where they both agree:

OEO(E)
   EO(E)

EOE(O)
   OE(O)

EEE(E)
   EE(E)

OOO(O)
   OO(O)

We can see right there they're the symmetrical ones in all 8 possible combinations. The remaining 4 will be the asymmetrical types, where both do not agree on which the next outcome could be, and cancel each other out, thus No Bets:

OEE(O)
   EE(E)

EOO(E)
   OO(O)

OEE(O)
   EE(E)

EOO(E)
   OO(O)

Later...

[psimoes]

Now with L1A added to antagonize L3F and L2F and see which patterns remain "bettable". For L3A+L2A, L1F is selected.

OEO(E)
   EO(E)
     O(E)

EOE(O)
   OE(O)
      E(O)

EEE(E)
   EE(E)
     E(O)

OOO(O)
   OO(O)
      O(E)

For the next, L1 is superfluous, since L3+L2 don't match already, invalidating all bets.

OEE(O)
   EE(E)
     
EOO(E)
   OO(O)

OEE(O)
   EE(E)

EOO(E)
   OO(O)

So an antagonizing L1A or L1F limits our bets even more. All we have left are the series of chops and they must go for at least 3 spins before we bet for/against. There we go again, waiting for specific events before we bet LOL. At least we're trying to justify our bet selections with a little more reasoning...

[psimoes]

Now, do we really need L1A or L1F? If we select L3F+L2A, as on the following examples, we avoid all symmetrical runs such as streaks and chops and bet only when the outcomes are seemingly chaotic, like the 4 asymmetrical patterns where L3 and L2 agree:

OEO(E)
   EO(O)
   
EOE(O)
   OE(E)     

EEE(E)
   EE(O)
     
OOO(O)
   OO(E)
     
OEE(O)
   EE(O)
     
EOO(E)
   OO(E)

OEE(O)
   EE(O)

EOO(E)
   OO(E)

In the pdf file with the session I tested there were more LLLLLL in a row, choosing the above option.

So, what do you guys think is more favourable?

Rewster88 and RFMAXX, I hope your answers have been adressed by now, but if you, or anyone else, still need more explanations don't hesitate to ask.

Edit - currently testing T2 today's SBWB permanences.

[Nickmsi]

Hello

Here's a thought.

I have done a lot of testing with EC bets and found what is called the Arcsine Principal or something to that effect.  I am not a math person.

Years ago, I ran across this principal in a thread by Beretta and just last year Bayes did a thread on it.

It basically says that if you are betting a "trend" then the probability that the "trend" will continue is slightly higher than betting that it won't.  I don't know the degree of probability.

So instead of betting "O" against EEE you would bet that another "E" is spun.  Then if the streak continues you would win each and every one of them until the trend is broken. 

If you have "OEO" you would bet an "E" so that the chopping streak will continue.

In other words, bet with the trend.

Just something to consider.

Cheers

Nick

[psimoes]

Nickmsi, thank you for a most interesting contribution! I'll read about it later. I'm not a "math person" either despite posting a lot of numbers LOL. Just find it fascinating among other areas. Which I was more educated on the matter.

That validates L3F+L2F; and L2F alone as my preferred choice for betting on ECs. Too bad it still loses to adverse runs of two, so I'm always looking for ways to improve it.

And I guess that validates betting for the same as last 12 outcomes too, unless that Arcsine Principal is only valid for higher resolution analysis i.e. spin by spin. If it's valid for longer cycles as well, it would confirm the fractal properties of chaos. That might explain why stiff patterns in betting methods get counter-intuitive after a while. Like the human brain detects the longer cycle trends and predicts they'll reflect in the shorter cycles. Sorry I digress.

I'm not a "pattern breaker" myself either, (as L2F proves). Last comparison, though, favoured betting against the trend. Maybe I was being too ambitious.

Anyway already started testing L3+L2+L1 so I'll finish it and post it right away.

In the meanwhile previous results are attached. Comparisons later.

Cheers.